Modulate

Voice-native models for complete understanding of any conversation

Velma Our flagship model with dynamic behaviors and all conversation signals

Transcription Fastest and cheapest multilingual transcription with diarization

Deepfake Detection #1 on Hugging Face for catching synthetic & cloned voices

PII/PHI Redaction Redacts 100+ sensitive entities from transcript and audio

Music & Speech Detection Separates music, speech, and silence at 0.2s frames

Medical Transcription Precision tuned for clinical and medical terminology

AI Music Detection Detects AI-generated music and instruments at 4s frames

Language Detection Identifies 100+ languages at blazing speed

Emotion Detection Reads emotional signals directly from voice, not just words

Accent Identification Identifies speaker accents and dialects across languages

Voice Match Tells whether two recordings come from the same speaker

Pricing Docs

Welcome to Modulate's Developer APIs!